PROJECT ABSTRACT The proposed study will help to establish methods for rapid detection of heteroresistant Mycobacterium tuberculosis (MTB) infections among HIV-infected TB patients. Heteroresistant MTB infections involve both drug-resistant and drug-susceptible types of MTB. A person with heteroresistant MTB infection is more likely to develop fully drug-resistant TB, which can cause poor treatment outcomes, increasing morbidity and mortality. People living with HIV appear to be especially at high risk for acquiring heteroresistant TB infections, and TB remains the leading cause of death among HIV-infected persons in resource poor countries, including Botswana. Recent developments in sequencing technology (deep sequencing) could be a powerful tool for rapidly detecting heteroresistant MTB infections. However, no study has systematically evaluated deep sequencing for detecting heteroresistance among TB patients in a clinical setting. Improved understanding of heteroresistance will contribute to significant improvements in TB morbidity and mortality and advance scientific understanding of TB. We propose to conduct an exploratory substudy of TB/HIV co-infected patients enrolled in an ongoing population-based TB research study in Botswana. We will identify and enroll 200 newly diagnosed TB patients on first-line therapy. We will collect sputum samples at enrollment and after 2 months of treatment. We will use deep sequencing of MTB DNA to look for heteroresistance. There are two specific aims to our proposal. Aim 1. We will determine whether direct sequencing from sputum specimen increases the detection of heteroresistant MTB infections compared to sequencing done on MTB culture samples. Aim 2. We will determine whether heteroresistant MTB infection status is predictive of response to TB treatment among patients on first-line TB therapy. This proposed research will be the first to systematically evaluate the use of deep sequencing to detect heteroresistant MTB infections, potentially improving clinical practice by allowing for rapid, early detection of heteroresistant infections and informing treatment decisions accordingly.